U.S. patent number 9,857,313 [Application Number 15/515,227] was granted by the patent office on 2018-01-02 for method and system for inspecting wafers for electronics, optics or optoelectronics.
This patent grant is currently assigned to UNITY SEMICONDUCTOR. The grantee listed for this patent is UNITY SEMICONDUCTOR. Invention is credited to Mayeul Durand De Gevigney, Philippe Gastaldo.
United States Patent |
9,857,313 |
Durand De Gevigney , et
al. |
January 2, 2018 |
Method and system for inspecting wafers for electronics, optics or
optoelectronics
Abstract
A method for inspecting a wafer, includes: rotating the wafer
about an axis of the wafer, emitting from a light source, two pairs
of incident coherent light beams, each pair forming, at the
intersection between the two beams, a measurement volume, a portion
of the main wafer surface passing through each of the measurement
volumes during the rotation, collecting a light beam scattered by
the wafer surface, capturing the collected light and emitting an
electrical signal representing the variation in the collected light
intensity, detecting in the signal, a frequency, being the time
signature of a defect through a respective measurement volume, for
each detected signature, determining a visibility parameter, on the
basis of the visibility determined, obtaining an item of
information on the size of the defect, and cross-checking the items
of information to determine the size of the defect.
Inventors: |
Durand De Gevigney; Mayeul
(Meylan, FR), Gastaldo; Philippe (Pontcharra,
FR) |
Applicant: |
Name |
City |
State |
Country |
Type |
UNITY SEMICONDUCTOR |
Monbonnot-Saint-Martin |
N/A |
FR |
|
|
Assignee: |
UNITY SEMICONDUCTOR
(Montbonnot-Saint-Martin, FR)
|
Family
ID: |
51866250 |
Appl.
No.: |
15/515,227 |
Filed: |
September 29, 2015 |
PCT
Filed: |
September 29, 2015 |
PCT No.: |
PCT/EP2015/072364 |
371(c)(1),(2),(4) Date: |
March 29, 2017 |
PCT
Pub. No.: |
WO2016/050735 |
PCT
Pub. Date: |
April 07, 2016 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20170219496 A1 |
Aug 3, 2017 |
|
Foreign Application Priority Data
|
|
|
|
|
Sep 29, 2014 [FR] |
|
|
14 59172 |
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G01N
21/8806 (20130101); G01N 21/9505 (20130101); G01N
21/8851 (20130101); G01N 21/9503 (20130101); G01B
11/2441 (20130101); G01N 21/9501 (20130101); G01M
11/331 (20130101); G01N 2021/8874 (20130101) |
Current International
Class: |
G01N
21/00 (20060101); G01N 21/95 (20060101); G01N
21/88 (20060101) |
Field of
Search: |
;356/237.5 |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
Other References
International Search Report from International Patent Application
No. PCT/EP2015/072364, dated Nov. 27, 2015. cited by applicant
.
Written Opinion from International Patent Application No.
PCT/EP2015/072364, dated Nov. 27, 2015. cited by applicant .
Farmer, W.M., "Measurement of Particle Size, Number Density, and
Velocity Using a Laser Interferometer" Applied Optics (1972),
11(11), p. 2603-2612. (Abstract only). cited by applicant .
Tanner, L. H., "A Study of Fringe Clarity in Laser Interferometry
and Holography" Journal of Physics E: Scientific Instruments
(1968), 1(5). (Abstract Only). cited by applicant.
|
Primary Examiner: Punnoose; Roy M
Attorney, Agent or Firm: Greer, Burns & Crain, Ltd.
Claims
The invention claimed is:
1. A method for the inspection of a wafer for electronics, optics
or optoelectronics, comprising: rotating the wafer about an axis of
symmetry perpendicular to a main surface of said wafer; emitting,
from at least one light source coupled with an interferometric
device, at least two pairs of incident coherent light beams, each
pair being arranged in order to form, at the intersection between
the two beams, a respective measurement volume containing
interference fringes having an inter-fringe distance different from
that of another measurement volume; at least a portion of the main
surface of the wafer passing through each of said measurement
volumes during the rotation of the wafer; collecting at least a
portion of the light scattered by said part of the surface of the
wafer; capturing the collected light and emitting an electrical
signal representing the variation in the light intensity of the
collected light as a function of time; detecting, in said signal, a
frequency component in the variation of the light intensity of said
collected light, said frequency being the time signature of the
passage of a defect through a respective measurement volume; for
each detected signature, determining a parameter, called visibility
of the defect, dependent on the inter-fringe distance of the
respective measurement volume and the size of the defect and given
by the following formula, determined on the basis of a Doppler
signal due to a defect passing through the measurement volume and
expressed in the form of an electrical voltage as a function of
time: .times. ##EQU00004## where Imax and Imin define the minimum
electrical voltage and the maximum electrical voltage defining the
peak of said Doppler signal and Offset is that between the mean
value of the Doppler signal and an axis corresponding to a zero
electrical voltage; on the basis of the visibility determined for
each measurement volume, obtaining a respective item of information
on the size of said defect; and cross-checking the information
obtained for each measurement volume in order to determine the size
of the defect.
2. The method according to claim 1, in which obtaining an item of
information on the size of the defect comprises: calculating the
visibility of the defect in each measurement volume, for each
measurement volume, on the basis of a visibility reference curve as
a function of the size of the defect for the respective
inter-fringe distance, determining one or more possible sizes for
the defect.
3. The method according to claim 1, said method comprising
filtering the signal with a band-pass filter the pass-band of which
incorporates the Doppler frequency associated with each measurement
volume.
4. The method according to claim 1, in which said measurement
volumes are at least partially superimposed.
5. The method according to claim 1, in which said measurement
volumes follow one another along the path of rotation of the
wafer.
6. The method according to claim 1, comprising a radial movement of
said measurement volumes with respect to the wafer.
7. The method according to claim 1, in which the fringes of each
measurement volume are oriented transversally to the path of
rotation of the wafer.
8. The method according to claim 1, in which the interferometric
device is an integrated optical device comprising a light guide the
input of which is coupled with the light source and which is
divided into two pairs of branches, the output of which is oriented
in order to form a respective measurement volume at the
intersection of the two beams of each pair.
9. The method according to claim 1, in which the wafer is at least
partially transparent vis-a-vis the wavelength of the light source
and each measurement volume extends into a region of the wafer
having a thickness less than the thickness of said wafer.
10. A system for inspecting wafers for electronics, optics or
optoelectronics, comprising: a device for driving a wafer in
rotation about an axis of symmetry perpendicular to a main surface
of said wafer; at least one light source; at least one
interferometric device coupled with the light source in order to
divide the beam emitted by said source into two pairs of beams and
in order to form, at the intersection between two beams of each
pair, a respective measurement volume containing interference
fringes, having an inter-fringe distance that is different from
that of another measurement volume; a device for the collection of
at least a portion of the light scattered by the surface of the
wafer; a device for capturing the collected light configured in
order to emit an electrical signal representing the variation in
the light intensity of said collected light as a function of time;
a processing device configured in order to: detect, in said signal,
a frequency component in the variation of the intensity of said
collected light, said frequency being the time signature of the
passage of a defect through a respective measurement volume; for
each detected signature, determine a parameter, called visibility
of the defect, dependent on the inter-fringe distance of the
respective measurement volume and the size of the defect, and given
by the following formula, determined on the basis of a Doppler
signal due to a defect passing through the measurement volume and
expressed in the form of an electrical voltage as a function of
time: .times. ##EQU00005## where Imax and Imin define the minimum
electrical voltage and the maximum electrical voltage defining the
peak of said Doppler signal and Offset is that between the mean
value of the Doppler signal and an axis corresponding to a zero
electrical voltage; obtain, on the basis of the visibility
determined for each measurement volume, a respective item of
information on the size of said defect; and cross-check the
information obtained for each measurement volume in order to
determine the size of the defect.
11. The system according to claim 10, comprising a single light
source and a single interferometric device in order to form all the
measurement volumes.
12. The system according to claim 10, in which the interferometric
device is in the form of an integrated optical device comprising a
light guide the input of which is coupled with the light source and
which is divided into two pairs of branches, the output of which is
oriented in order to form a respective measurement volume at the
intersection of the two beams of each pair.
13. The system according to claim 10, also comprising an arm for
moving the interferometric device and the device for the collection
of scattered light in translational motion in a radial direction.
Description
BACKGROUND
The present invention relates to a method and a system for
inspecting wafers for electronics, optics or optoelectronics.
During the manufacture and use of wafers for electronics, optics or
optoelectronics, it is usual to carry out an inspection of the
surface of each wafer so as to detect any defects.
On account of the very small size of the defects to be detected, a
visual inspection by an operator is not sufficient.
Furthermore, the inspection is generally intended not only to
discover the presence or absence of defects, but also to provide
qualitative and/or quantitative information on said defects, such
as their location, their size and/or their nature, for example.
Inspection systems have thus been developed with a view to
detecting increasingly small defects and to provide all required
information on the nature, the size, the location, etc. of said
defects.
These systems must also allow a duration of inspection of each
wafer that is sufficiently short so as not to adversely affect
production speeds.
Document WO 2009/112704 describes a system for inspecting
semi-conductor wafers implementing Laser Doppler Velocimetry (LDV).
As shown in FIG. 1, this system comprises a light source 20 and an
interferometric device 30 coupled with the light source arranged
facing the surface S of the wafer 2 for inspection, which is
actuated by a movement. Said interferometric device comprises a
light guide the input of which is coupled with the light source and
comprising two branches for dividing the beam originating from the
light source into two incident beams. At the output of the light
guide, the two branches are oriented in relation to one another so
as to form, at the intersection between the two beams, a
measurement volume comprising a plurality of parallel fringes. The
system also comprises an optical fibre 40 arranged between the
surface of the wafer and a detection module 50, so as to guide the
light backscattered by the surface of the wafer towards the
detection module.
Document WO 02/39099 describes another system for inspecting
semi-conductor wafers relying on Laser Doppler Velocimetry.
The presence of a defect on the surface of the wafer is indicated,
when this defect crosses the interference fringes, by the
scattering of a Doppler burst measured by the detection module. A
Doppler burst is a signal that has a double frequency component: a
low-frequency component, forming the envelope of the signal,
corresponding to the mean light intensity scattered by the defect,
and a high-frequency component, corresponding to the Doppler
frequency containing the information on the velocity of the defect.
The Doppler frequency f.sub.D is linked to the velocity v of
movement of the defect in the direction perpendicular to the
interference fringes and to the distance .DELTA. between the
interference fringes (or inter-fringe distances) by the
relationship v=f*.DELTA..
FIG. 2 shows a Doppler burst due to a defect passing through the
interference zone, expressed in the form of an electrical voltage
(in Volts) at the output of the detection module as a function of
time.
On the basis of such a Doppler burst, it is possible to determine
the size of the defects detected on the surface of the wafer.
In this respect, reference may be made to the publication by W. M.
Farmer entitled "Measurement of Particle Size, Number Density, and
Velocity Using a Laser Interferometer", which presents a model of
the visibility of a particle as a function of the particle
size.
Thus, for a pattern of given interference fringes, the relationship
between the size of a defect compared to a sphere, which is defined
as the diameter of the sphere, and the visibility determined
according to the above formula, is given by a curve of the type
shown in FIG. 3.
It is noted that, for a visibility greater than 0.15, the curve of
FIG. 3 provides a unique defect size corresponding to a given
visibility value.
However, for a visibility less than 0.15, the curve shows
"bounces", indicating the fact that a single visibility value can
correspond to several defect sizes. Thus, in the example in FIG. 3,
a visibility of 0.1 corresponds to three radii of a sphere: 0.83
.mu.m, 1.12 .mu.m and 1.45 .mu.m.
In such a case, the problem then arises of determining, among these
different possible sizes, the actual size of the defect present on
the wafer.
In particular, this technique does not allow measurement of the
size of defects of very different sizes. In fact, as shown in FIG.
3, it is not possible to determine the size of defects having a
size larger than 0.9 .mu.m (corresponding to a visibility less than
0.15).
Now, the size of the defects capable of being detected on a wafer
extends over a wide range of dimensions, typically from a few tens
of nanometers to a few hundred micrometers.
Another drawback of the technique based on the curve in FIG. 3 is
that, for some defect sizes (for example a radius of 0.95 .mu.m),
the visibility is zero, i.e. no Doppler burst is produced.
Consequently, a defect of this size cannot be detected.
SUMMARY
A purpose of the invention is to overcome the aforementioned
drawbacks and to define a system and a method for inspecting wafers
that make it possible to detect all of the defects that may be
present on the wafer having a size larger than a few tens of
nanometers, and to determine with certainty the size of each
detected defect. This system and this method must also have
improved detection dynamics with respect to the existing systems
and methods, i.e. a greater capacity to detect a large number of
defects and to assess their size within a reduced timescale over a
broad range of defect sizes.
According to the invention, a method is proposed for inspecting a
wafer for electronics, optics or optoelectronics, comprising:
rotating the wafer about an axis of symmetry perpendicular to a
main surface of said wafer, emitting, from at least one light
source, at least two pairs of incident coherent light beams, each
pair being arranged in order to form, at the intersection between
the two beams, a respective measurement volume containing
interference fringes having an inter-fringe distance different from
that of another measurement volume,
at least a portion of the main surface of the wafer passing through
each of said measurement volumes during the rotation of the wafer,
collecting a light beam scattered by the surface of the wafer,
capturing the collected light and emitting an electrical signal
representing the variation in the light intensity of the collected
light as a function of time, detecting, in said signal, a frequency
component in the variation of the intensity of said collected
light, said frequency being the time signature of the passage of a
defect through a respective measurement volume, for each detected
signature, determining a parameter, called visibility of the
defect, dependent on the inter-fringe distance of the respective
measurement volume and the size of the defect and given by the
following formula, determined on the basis of a Doppler signal due
to a defect passing through the measurement volume and expressed in
the form of an electrical voltage as a function of time:
.times. ##EQU00001##
where Imax and Imin define the minimum electrical voltage and the
maximum electrical voltage defining the peak of said Doppler signal
and Offset is that between the mean value of the Doppler signal and
an axis corresponding to a zero electrical voltage, on the basis of
the visibility determined for each measurement volume, obtaining a
respective item of information on the size of said defect,
cross-checking the items of information obtained for each
measurement volume in order to determine the size of the
defect.
Particularly advantageously, obtaining an item of information on
the size of the defect comprises: calculating the visibility of the
defect in each measurement volume, for each measurement volume, on
the basis of a visibility reference curve as a function of the size
of the defect for the respective inter-fringe distance, determining
one or more possible sizes for the defect.
Preferably, said method comprises the filtering of the signal with
a band-pass filter the pass-band of which incorporates the Doppler
frequency associated with each measurement volume.
According to an embodiment, said measurement volumes are at least
partially superimposed.
According to another embodiment, said measurement volumes follow
one another along the path of rotation of the wafer.
Particularly advantageously, the method also comprises a radial
movement of said measurement volumes with respect to the wafer.
As a general rule, the fringes of each measurement volume are
oriented transversally to the path of rotation of the wafer.
According to a particularly advantageous embodiment, the
interferometric device is an integrated optical device comprising a
light guide the input of which is coupled with the light source and
which is divided into two pairs of branches, the output of which is
oriented in order to form a respective measurement volume at the
intersection of the two beams of each pair.
According to an embodiment of the invention, the wafer is at least
partially transparent vis-a-vis the wavelength of the light source
and each measurement volume extends into a region of the wafer
having a thickness less than the thickness of said wafer.
A further subject relates to a system for inspecting wafers for
electronics, optics or optoelectronics, comprising: a device for
driving a wafer in rotation about an axis of symmetry perpendicular
to a main surface of said wafer, at least one light source, at
least one interferometric device coupled with the light source in
order to divide the beam emitted by said source into two beams and
in order to form, at the intersection between the two beams, a
respective measurement volume containing interference fringes,
having an inter-fringe distance that is different from that of
another measurement volume, a device for the collection of light
scattered by the wafer, a device for capturing the collected light
configured in order to emit an electrical signal representing the
variation in the light intensity of said collected light as a
function of time, a processing device configured in order to:
detect, in said signal, a frequency component in the variation of
the intensity of said collected light, said frequency being the
time signature of the passage of a defect through a respective
measurement volume, for each detected signature, determine a
parameter, called visibility of the defect, dependent on the
inter-fringe distance of the respective measurement volume and the
size of the defect, and given by the following formula, determined
on the basis of a Doppler signal due to a defect passing through
the measurement volume and expressed in the form of an electrical
voltage as a function of time:
.times. ##EQU00002##
where Imax and Imin define the minimum electrical voltage and the
maximum electrical voltage defining the peak of said Doppler signal
and Offset is that between the mean value of the Doppler signal and
an axis corresponding to a zero electrical voltage. obtain, on the
basis of the visibility determined for each measurement volume, a
respective item of information on the size of said defect,
cross-check the information obtained for each measurement volume in
order to determine the size of the defect.
According to an advantageous embodiment, said system comprises a
single light source and a single interferometric device in order to
form all the measurement volumes.
According to a preferred embodiment, in which the interferometric
device is in the form of an integrated optical device comprising a
light guide the input of which is coupled with the light source and
which is divided into two pairs of branches, the output of which is
oriented in order to form a respective measurement volume at the
intersection of the two beams of each pair.
Particularly advantageously, the system also comprises an arm for
moving the interferometric device and the device for the collection
of the scattered beam in translational motion in a radial
direction.
BRIEF DESCRIPTION OF THE DRAWINGS
Other characteristics and advantages will become apparent from the
detailed description that follows, with reference to the attached
drawings in which:
FIG. 1 is a schematic diagram of an inspection system based on
laser Doppler velocimetry, as described in the document WO
2009/112704,
FIG. 2 shows an example of a Doppler burst,
FIG. 3 is a diagram showing the visibility (variable without units)
of a defect compared to a sphere as a function of its size (in the
case in question the radius of the sphere expressed in
micrometers),
FIG. 4 is a schematic diagram of the inspection system according to
an embodiment of the invention,
FIGS. 5A and 5B are schematic drawings of the interferometric
device according to two embodiments of the invention,
FIG. 6 is a diagram showing the visibility (without units) of a
defect compared to a sphere as a function of its size (radius of
the sphere in .mu.m), for an inspection system according to the
invention,
FIG. 7 shows the principle of the detection of defects implementing
a number N greater than or equal to two measurement volumes.
In the interests of clarity, the figures are not necessarily to
scale.
DETAILED DESCRIPTION
The present invention relates to any wafer intended for use in the
field of electronics, optics or optoelectronics. In particular, the
wafer can comprise at least one of the following materials: Si, Ge,
GaN, SiC, glass, quartz, sapphire, GaAs (non-limitative list).
Furthermore, the material of the wafer may or may not be partially
transparent at the wavelength of the light source of the inspection
system. In fact, according to a particularly advantageous
embodiment that will be described in detail below, the inspection
system provides a controlled depth of field, making it possible to
control the position of the measurement volume with respect to the
wafer, by arranging for the region of the wafer in which the
measurement volume extends to have a thickness that is less than
the thickness of the wafer. In this way it is ensured that the
detected defects are located on the surface for inspection or its
close neighbourhood, and not on the opposite surface.
In order to allow the unambiguous determination of the size of a
detected defect and to make each defect visible regardless of its
size within a range of a few tens of nanometers to a few hundred
micrometers, the invention proposes to form at least two
measurement volumes containing interference fringes and each having
a different inter-fringe distance.
The measurement volumes are arranged with respect to one another
such that a defect of the wafer passes into each of the measurement
volumes and generates, if appropriate, a respective Doppler
burst.
FIG. 4 is a schematic diagram of an inspection system 1 utilizing
such measurement volumes.
The system comprises a support 10 intended to receive a wafer 2 for
inspection and to drive it in rotation about an axis of symmetry X
perpendicular to a main surface S of said wafer. Generally, the
wafer has a circular shape but the invention is applicable to any
other shape.
The wafer 2 is held on the support 10 by any appropriate means,
such as electrostatic means, mechanical means, etc.
The mechanism for rotating the support is known per se and
therefore will not be described in detail.
The support 10 comprises one or more encoders (not shown) making it
possible to know the angular position of the wafer at any
moment.
The inspection system 1 also comprises a light source 20.
The light source 20 is typically a distributed feedback (DFB)
laser.
The light source is coupled with an interferometric device 30 that
will be described in detail with reference to FIG. 5.
The interferometric device 30 is designed in order to form at least
two measurement volumes (only one of which is shown
diagrammatically in FIG. 4 under the reference V) having different
inter-fringe distances. These measurement volumes can be totally or
partially within a single space. As will be explained below with
reference to FIG. 6, the inter-fringe distances are chosen so that
the visibility curves associated with each of these measurement
volumes are sufficiently different from one another so that a
defect that is not visible in one of the measurement volumes is
visible in the other measurement volume, and in order to remove any
ambiguities relating to the size of the detected defects.
It can be envisaged for the inspection system to comprise several
interferometric devices each coupled to a light source in order to
form a respective measurement volume, but this embodiment is less
advantageous in terms of bulk and cost. As a result, preferably,
the inspection system comprises a single light source and a single
interferometric device suitable for forming the different
measurement volumes.
The inspection system comprises in addition a device 40 for
collection of the light backscattered by the surface of the wafer.
This device 40 can comprise an optical fibre, preferably with a
large core diameter (i.e. typically between 100 and 1000 .mu.m in
diameter), the input of which is arranged facing the surface of the
wafer, close to the measurement volumes, and the output of which is
coupled with a device 50 for capturing collected light in order to
emit an electrical signal representing the variation in the light
intensity of the collected light as a function of time. Said device
50 typically comprises a photodetector.
Preferably, the interferometric device 30 and the device 40 for the
collection of the backscattered light are firmly fixed together. In
fact, the input of the collection device 40 must be positioned in
an appropriate manner with respect to the measurement volumes in
order to receive the light backscattered by the wafer.
Finally, the inspection system 1 comprises a processing device 60
configured in order to detect, in said signal, a frequency
component corresponding to the Doppler frequency.
The processing device 60 is advantageously coupled with an
interface (not shown) making it possible for a user to access the
results so as in particular to display them, record them and/or
print them.
In the case where the wafer is at least partially transparent
vis-a-vis the wavelength of the light source, arrangements are made
so that the region in which each measurement volume extends has a
thickness less than that of the wafer. The thickness of said region
is preferably less than or equal to 90% of the thickness of the
wafer. For example, for a wafer of 500 .mu.m to 1 mm in thickness,
arrangements are made so that the measurement volumes extend into a
region of the wafer having a thickness less than or equal to 100
.mu.m. The dimension of the measurement volume is characteristic of
the interferometric device and is defined by the angle between the
two branches of the light guide in which the light beam emitted by
the source propagates and by the numerical aperture of said
branches.
It will be noted in this respect that the inspection systems
currently available on the market do not allow satisfactory
inspection of transparent wafers.
In fact, in the case of the systems based on the dark field
inspection technique, the incident beam passes through the
thickness of the wafer and any defect, whether present on the
surface to be inspected, on the opposite surface or in the
thickness of the substrate, generates a scattered light. It is
therefore impossible, with such a system, to know whether each
detected defect is located on the surface for inspection or
not.
Furthermore, the company KLA-Tencor proposes a system for
inspecting transparent wafers called Candela.TM., of the dark field
illumination and confocal laser detection type. However, this
system is particularly difficult to focus due to the accuracy of
positioning required for the confocal detection, and therefore does
not provide repeatable results.
The system implemented in the invention overcomes the constraints
associated with the dark field technique and with the confocal
detection technique by detecting the defects by using a frequency
signal, which can only be emitted by defects passing through a
measurement volume. In such a system, the positioning of the
interferometric device must therefore be adjusted accurately with
respect to the surface of the wafer for inspection, but the device
for the collection of the backscattered light does not require an
equally high positioning accuracy since the restriction of the
measurement volume, and thus the detection, is carried out via the
Doppler frequency.
Furthermore, in order to inspect transparent wafers, an integrated
optical device such as that described below is preferably chosen in
order to produce the interferometric device. Such a device in fact
makes it possible to control the depth of field of the inspection
system. On the other hand, measurement of the size by a visibility
calculation is independent of the position of the defect in the
measurement volume.
In order to inspect a wafer, said wafer 2 is put in place on the
support 10 and the support is driven in rotation at a controlled
angular velocity go. By means of the encoders present on the
support 10, the angular position of a given point of the wafer is
known at each moment. The velocity of rotation of the wafer is
typically of the order of 5000 rpm.
In the inspection system 1, the interferometric device 30 is
arranged facing a main surface of the wafer 2, on an arm (not
shown) suitable for moving said device 30 in a radial direction.
Thus, taking account of the rotation of the wafer, it is possible
to successively sweep the entire surface of the wafer with the
measurement volumes by moving the interferometric device radially
in translation as well as the device for collection of the
backscattered light.
The two measurement volumes are formed on the same side of the
wafer, in order to ensure that a defect passes through all the
measurement volumes. The interference fringes of each measurement
volume are oriented transversally with respect to the path of
rotation of the wafer, so as to be passed through by the defects.
The inclination between the fringes and the path of rotation of the
wafer can be perpendicular or according to another non-zero
angle.
According to the aforementioned principle of the method of W. M.
Farmer, for each measurement volume the visibility of a detected
defect is calculated by the formula:
.times. ##EQU00003##
where Imax and Imin (in V) define the minimum electrical voltage
and the maximum electrical voltage defining the peak of the Doppler
burst and Offset (in V) is that between the mean value of the
signal and the x-axis corresponding to a zero electrical voltage
(cf FIG. 2). This offset, which does not appear in the formula of
W. M. Farmer, is linked to the measurement conditions, and takes
account of the fact that even in the absence of a defect, a small
quantity of light scattered by the surface can be detected.
In addition, a plurality of reference curves of the type of that in
FIG. 3 are stored in a memory of the processing device, each
reference curve defining the visibility of a defect in a respective
measurement volume as a function of the size of the defect.
In an embodiment of the invention, the measurement volumes follow
one another along the path of rotation of the wafer, at the same
radial distance from the axis of rotation of the wafer. Thus, the
defects pass successively through the different measurement volumes
during the rotation of the wafer.
According to another embodiment of the invention, the measurement
volumes are at least partially superimposed. In fact, subject to
implementing band-pass filtering integrating the Doppler frequency
associated with each inter-fringe distance and therefore with each
measurement volume, the signal emitted by the photodetector
contains only the information linked to these measurement volumes
and makes it possible for them to be distinguished. By
"integrating" is meant here that the pass-band of the filter
comprises the Doppler frequency and a small frequency range around
this Doppler frequency.
FIGS. 5A and 5B are schematic drawings of the two embodiments of an
interferometric device making it possible to form two measurement
volumes containing interference fringes, each having a different
inter-fringe distance. In the case of FIG. 5A, the measurement
volumes are adjacent; in the case of FIG. 5B, the measurement
volumes are at least partially superimposed.
This device 30 comprises a light guide 31 the input 32 of which is
coupled with the light source 20 and comprising two symmetrical
main branches 33, 34 for dividing the beam originating from the
light source into two incident beams.
Each branch 33, 34 is itself divided into two symmetrical secondary
branches, respectively 33a, 33b and 34a, 34b.
At its end, each secondary branch has an expanded portion intended
to widen the beam while retaining its Gaussian profile.
At the output of the light guide, the secondary branches of each
pair are oriented in relation to one another so as to form, at the
intersection between the two beams, a measurement volume containing
parallel interference fringes. As shown diagrammatically in FIG.
5A, the pair 33a, 33b forms a measurement volume the inter-fringe
distance of which has a value .DELTA.1 and the pair 34a, 34b forms
a measurement volume the inter-fringe distance of which has a value
.DELTA.2 different from .DELTA.1.
The device in FIG. 5B follows the same principle as that in FIG.
5A, but the fringes have not been shown in order to simplify the
figure. In this embodiment, the different branches are
symmetrically interleaved so that the measurement volumes created
at the output of said branches substantially coincide.
Particularly advantageously, the interferometric device is in the
form of an integrated sensor constituted by a single piece and
ensuring both the separation of the beam emitted by the light
source and the transmission of the pairs of branches of the beam in
order to form the interference volumes at the output of the sensor.
It is noted that an integrated optical device is an optical device
produced by microelectronic techniques.
The article "Integrated Laser Doppler Velocimeter for Fluid
Velocity and Wall Friction Measurements" by P. Lemaitre-Auger et
al. describes such a sensor (which in this case has a single main
branch and two secondary branches, so as to form a single
measurement volume). Such a device is produced in particular by the
company A2 Photonic Sensors and marketed under the reference
i-LDA.TM..
The same method of manufacture as that described in the
aforementioned article can be implemented in order to integrate
several light guides within the sensor in order to form several
measurement volumes.
By way of example, the integrated optical device can be produced by
ion exchange on a glass substrate. This process generally
comprises: providing a glass substrate, depositing a metallic
masking layer onto said glass substrate, depositing a polymer layer
onto the metallic layer, transferring by photolithography a pattern
defining the shape of the light guide onto the polymer layer,
chemical etching of the metallic masking layer using a chemical
process in the zones left exposed by the polymer mask, removing the
polymer mask, immersing the substrate covered with the etched
metallic masking layer in an ion bath (for example a potassium
nitrate bath), exchanging ions present in the bath (for example
potassium ions) and ions contained in the glass (for example sodium
ions) through zones of the substrate that are not covered by the
metallic masking layer, the latter blocking the passage of the
ions.
On account of the difference in size between the ions present in
the bath and the ions present in the glass, the ion exchange
generates local mechanical stresses in the glass substrate which
increase the refractive index of the glass. The aforementioned
optical waveguide is obtained in this way.
The metallic masking layer is then removed and optionally a
protective layer, for example of SiO.sub.2, is deposited. Finally,
the edges of the substrate are cut out and they are finely
polished.
There are other processes for the manufacture of integrated optical
devices and a person skilled in the art may choose from the
microelectronic technologies at their disposal in order to design
the integrated optical device.
Optionally, the optical device may also be combined with an optical
fibre making it possible to collect the backscattered light.
An advantage of this integrated device is its robustness and its
stability. In particular, unlike a system produced by other
technologies such as micro-optics or optical fibres, the compact
nature of the integrated device and the integration of the various
components means that it is not sensitive to vibration and
temperature gradients.
Advantageously, when it is desired to inspect wafers that are least
partially
transparent with respect to the wavelength of the light source,
arrangements are made to ensure that the thickness of the region of
the wafer in which the measurement volume extends is less than the
thickness of the wafer (this region including a portion of the
surface area to be inspected). The thickness of said region is
preferably less than or equal 25 to 90% of the thickness of the
wafer. For example, for a wafer of 500 .mu.m to 1 mm in thickness,
arrangements are made so that the measurement volume extends into a
region of the wafer having a thickness less than or equal to 100
.mu.m. The dimension of the measurement volume is characteristic of
the interferometric device and is defined by the angle between the
two branches of the light guide in which the light beam emitted by
the source propagates and by the numerical aperture of said
branches. These characteristics are thus set during the manufacture
of the integrated optical device, which makes it possible to ensure
good control of the performances of the system during its mass
production.
Thus, it is possible to limit this measurement volume to the
surface of the wafer or to a region of the neighbourhood of said
surface. In this way it is ensured that the detected defects are
located on the surface to be inspected or its close neighbourhood,
and not on the opposite surface of the wafer.
An integrated optical device has an additional advantage in this
context, given that its stability makes it possible to avoid a
depth-of-field error. The control of the depth of field permitted
by the integrated device thus facilitates the inspection of
transparent wafers by laser Doppler velocimetry
It will be noted that by contrast, the control of the depth of
field assumes a lesser importance for inspecting an opaque wafer,
given that, since the measurement volume does not penetrate into
the thickness of such a wafer, it is sufficient for a portion of
the surface of the wafer to pass through the measurement volume in
order to allow the inspection of said surface.
As stated in the aforementioned article, the inter-fringe distance
depends on the wavelength of the light source, the optical index of
the lightguide and the angle between the two secondary branches.
For a given wavelength of the light source, the inter-fringe
distance is thus fixed during the manufacture of the integrated
optical device.
FIG. 6 shows two examples of visibility curves as a function of the
size of the defect for two different inter-fringe distances.
The curve (a) corresponds substantially to the curve in FIG. 3.
It is noted that curve (b) has fewer "bounces" corresponding to
zero visibility than curve (a), and that said points of zero
visibility do not coincide with the points of zero visibility of
curve (a).
Thus, if a defect has zero visibility in the measurement volume
corresponding to curve (a), it cannot be detected via curve (a); on
the other hand, as it has a visibility that is non-zero in the
measurement volume corresponding to curve (b), it can be detected
via said curve (b).
For example, a defect of radius 1.7 .mu.m has zero visibility on
curve (a) but a visibility of around 0.22 on curve (b) and thus
will be detectable on curve (b).
Furthermore, this offset of the visibility curves makes it possible
to remove the ambiguities on the size of the detected defects by
cross-checking the information supplied by both curves. In fact, by
choosing visibility curves that are sufficiently distant from one
another, a visibility corresponding to several possible defect
sizes on one of the curves will only correspond to one defect size
on the other curve.
For example, a defect of 1.5 .mu.m has a visibility of 0.07 on
curve (a). Now, on curve (a), a visibility of 0.07 corresponds to
four sizes of defect. 0.8 .mu.m, 1 .mu.m, 1.5 .mu.m and 2 .mu.m;
this single visibility value therefore does not allow a conclusion
to be drawn on the size of the detected defect. On the other hand,
on curve (b), this same defect of radius 1.5 .mu.m has a visibility
of 0.33. As a result, knowledge of the visibilities of 0.07 and
0.33 makes it possible to conclude unambiguously that the radius of
the detected defect is 1.5 .mu.m.
A person skilled in the art is able to determine the inter-fringe
distance of each measurement volume in order to allow the
determination of the size of a defect throughout the entire extent
of the size to be detected. On the basis of curves of the type of
that in FIG. 3, which can be obtained by simulation according to
the method described by W. M. Farmer, a person skilled in the art
will seek to have a sufficiently high visibility for each
measurement volume and avoid the case where the combination of the
items of information collected based on each measurement volume may
correspond to several sizes of defects.
Although up to this point embodiments have been described with two
measurement volumes having different inter-fringe distances, the
invention can more generally be implemented with an integer N
greater than or equal to two measurement volumes each having a
specific inter-fringe distance. With three measurement volumes or
more, in fact the accuracy of determination of the size of the
defects will be further increased.
FIG. 7 is a logic diagram showing the sequence of detection of
defects with a number N of measurement volumes greater than
two.
The light source 20 is coupled to the input of the interferometric
device which comprises N pairs of secondary branches, each pair
being designed so as to have a different inter-fringe distance
.DELTA.1, .DELTA.2, . . . , .DELTA.N.
The device 40 for the collection of the backscattered light is
common to the set of measurement volumes, as is the capturing
device 50 and the processing device 60.
In the processing device 60, the signal supplied by the capturing
device 50 is filtered by N band-pass filters each having a
different pass-band B1, B2, . . . BN, incorporating the Doppler
frequency associated with a respective inter-fringe distance
.DELTA.1, .DELTA.2, . . . , .DELTA.N.
As a result, the filtered signal provides N items of information
S1, S2, . . . SN on the size of the detected defects. In the case
where a defect is not visible in one of the measurement volumes,
the corresponding information is an absence of a defect. In the
case where a defect has a visibility associated with different
possible sizes, the corresponding information is the set of
possible sizes.
The set of information S1, S2, . . . , SN is then combined in order
to allow, by cross-checking, unambiguous determination of the size
of each detected defect (step shown diagrammatically by block
C).
The detection device then provides a report R on the detected
defects, indicating the size and the position of each defect.
REFERENCES
WO 2009/112704 WO 02/39099 Measurement of Particle Size, Number
Density, and Velocity Using a Laser Interferometer, W. M. Farmer,
Applied Optics, Vol. 11, No. 11, November 1972, pp. 2603-2612
Integrated Laser Doppler Velocimeter for Fluid Velocity and Wall
Friction Measurements, P. Lemaitre-Auger, A. Cartellier, P. Benech
and Schanen Duport, Sensors, 2002, Proceedings of IEEE (Vol: 1),
pp. 78-82.
* * * * *